diff options
author | James Almer <jamrial@gmail.com> | 2015-01-12 21:32:24 +0300 |
---|---|---|
committer | James Almer <jamrial@gmail.com> | 2015-01-12 21:40:03 +0300 |
commit | 59ac93f6af3e4e148e744e5820a1d56fadd635d6 (patch) | |
tree | 03f2ec5613722221121cdd5fbaf844eb46b4fd74 /libswresample | |
parent | e9f8780381feb7a62939018149e5153dfffaf921 (diff) |
x86/swr: add SSE/AVX unpack_6ch functions
int32/float only
Reviewed-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: James Almer <jamrial@gmail.com>
Diffstat (limited to 'libswresample')
-rw-r--r-- | libswresample/x86/audio_convert.asm | 89 | ||||
-rw-r--r-- | libswresample/x86/audio_convert_init.c | 16 |
2 files changed, 105 insertions, 0 deletions
diff --git a/libswresample/x86/audio_convert.asm b/libswresample/x86/audio_convert.asm index 07d74587b8..1617e0b645 100644 --- a/libswresample/x86/audio_convert.asm +++ b/libswresample/x86/audio_convert.asm @@ -305,6 +305,79 @@ pack_6ch_%2_to_%1_u_int %+ SUFFIX %endif %endmacro +%macro UNPACK_6CH 5-7 +cglobal unpack_6ch_%2_to_%1_%3, 2, 8, 8, dst, src, dst1, dst2, dst3, dst4, dst5, len +%if ARCH_X86_64 + mov lend, r2d +%else + %define lend dword r2m +%endif + mov dst1q, [dstq+1*gprsize] + mov dst2q, [dstq+2*gprsize] + mov dst3q, [dstq+3*gprsize] + mov dst4q, [dstq+4*gprsize] + mov dst5q, [dstq+5*gprsize] + mov dstq, [dstq] + mov srcq, [srcq] +%ifidn %3, a + test dstq, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX + test srcq, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX + test dst1q, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX + test dst2q, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX + test dst3q, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX + test dst4q, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX + test dst5q, mmsize-1 + jne unpack_6ch_%2_to_%1_u_int %+ SUFFIX +%else +unpack_6ch_%2_to_%1_u_int %+ SUFFIX +%endif + sub dst1q, dstq + sub dst2q, dstq + sub dst3q, dstq + sub dst4q, dstq + sub dst5q, dstq + %7 x,x,x,x,m7,x +.loop: + mov%3 m0, [srcq ] + mov%3 m1, [srcq+16] + mov%3 m2, [srcq+32] + mov%3 m3, [srcq+48] + mov%3 m4, [srcq+64] + mov%3 m5, [srcq+80] + + SBUTTERFLYPS 0, 3, 6 + SBUTTERFLYPS 1, 4, 6 + SBUTTERFLYPS 2, 5, 6 + SBUTTERFLYPS 0, 4, 6 + SBUTTERFLYPS 3, 2, 6 + SBUTTERFLYPS 1, 5, 6 + SWAP 1, 4 + SWAP 2, 3 + + %6 m0,m1,x,x,m7,m6 + %6 m2,m3,x,x,m7,m6 + %6 m4,m5,x,x,m7,m6 + + mov %+ %3 %+ ps [dstq ], m0 + mov %+ %3 %+ ps [dstq+dst1q], m1 + mov %+ %3 %+ ps [dstq+dst2q], m2 + mov %+ %3 %+ ps [dstq+dst3q], m3 + mov %+ %3 %+ ps [dstq+dst4q], m4 + mov %+ %3 %+ ps [dstq+dst5q], m5 + + add srcq, mmsize*6 + add dstq, mmsize + sub lend, mmsize/4 + jg .loop + REP_RET +%endmacro + %macro PACK_8CH 5-7 cglobal pack_8ch_%2_to_%1_%3, 2,10,10, ARCH_X86_32*32, dst, src, len, src1, src2, src3, src4, src5, src6, src7 mov dstq, [dstq] @@ -527,6 +600,9 @@ INIT_XMM sse PACK_6CH float, float, u, 2, 2, NOP_N, NOP_N PACK_6CH float, float, a, 2, 2, NOP_N, NOP_N +UNPACK_6CH float, float, u, 2, 2, NOP_N, NOP_N +UNPACK_6CH float, float, a, 2, 2, NOP_N, NOP_N + INIT_XMM sse2 CONV int32, int16, u, 2, 1, INT16_TO_INT32_N, NOP_N CONV int32, int16, a, 2, 1, INT16_TO_INT32_N, NOP_N @@ -583,6 +659,11 @@ PACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT PACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT PACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT +UNPACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT +UNPACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT +UNPACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT +UNPACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT + %if HAVE_ALIGNED_STACK PACK_8CH float, float, u, 2, 2, NOP_N, NOP_N PACK_8CH float, float, a, 2, 2, NOP_N, NOP_N @@ -606,11 +687,19 @@ INIT_XMM avx PACK_6CH float, float, u, 2, 2, NOP_N, NOP_N PACK_6CH float, float, a, 2, 2, NOP_N, NOP_N +UNPACK_6CH float, float, u, 2, 2, NOP_N, NOP_N +UNPACK_6CH float, float, a, 2, 2, NOP_N, NOP_N + PACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT PACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT PACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT PACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT +UNPACK_6CH float, int32, u, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT +UNPACK_6CH float, int32, a, 2, 2, INT32_TO_FLOAT_N, INT32_TO_FLOAT_INIT +UNPACK_6CH int32, float, u, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT +UNPACK_6CH int32, float, a, 2, 2, FLOAT_TO_INT32_N, FLOAT_TO_INT32_INIT + %if HAVE_ALIGNED_STACK PACK_8CH float, float, u, 2, 2, NOP_N, NOP_N PACK_8CH float, float, a, 2, 2, NOP_N, NOP_N diff --git a/libswresample/x86/audio_convert_init.c b/libswresample/x86/audio_convert_init.c index b38889fb3a..bd5f741b9a 100644 --- a/libswresample/x86/audio_convert_init.c +++ b/libswresample/x86/audio_convert_init.c @@ -31,6 +31,7 @@ PROTO4(_pack_2ch_) PROTO4(_pack_6ch_) PROTO4(_pack_8ch_) PROTO4(_unpack_2ch_) +PROTO4(_unpack_6ch_) av_cold void swri_audio_convert_init_x86(struct AudioConvert *ac, enum AVSampleFormat out_fmt, @@ -63,6 +64,9 @@ MULTI_CAPS_FUNC(SSE2, sse2) if(channels == 6) { if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) ac->simd_f = ff_pack_6ch_float_to_float_a_sse; + + if( out_fmt == AV_SAMPLE_FMT_FLTP && in_fmt == AV_SAMPLE_FMT_FLT || out_fmt == AV_SAMPLE_FMT_S32P && in_fmt == AV_SAMPLE_FMT_S32) + ac->simd_f = ff_unpack_6ch_float_to_float_a_sse; } } if(EXTERNAL_SSE2(mm_flags)) { @@ -116,6 +120,11 @@ MULTI_CAPS_FUNC(SSE2, sse2) ac->simd_f = ff_pack_6ch_int32_to_float_a_sse2; if( out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_FLTP) ac->simd_f = ff_pack_6ch_float_to_int32_a_sse2; + + if( out_fmt == AV_SAMPLE_FMT_FLTP && in_fmt == AV_SAMPLE_FMT_S32) + ac->simd_f = ff_unpack_6ch_int32_to_float_a_sse2; + if( out_fmt == AV_SAMPLE_FMT_S32P && in_fmt == AV_SAMPLE_FMT_FLT) + ac->simd_f = ff_unpack_6ch_float_to_int32_a_sse2; } if(HAVE_ALIGNED_STACK && channels == 8) { if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) @@ -146,6 +155,13 @@ MULTI_CAPS_FUNC(SSE2, sse2) ac->simd_f = ff_pack_6ch_int32_to_float_a_avx; if( out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_FLTP) ac->simd_f = ff_pack_6ch_float_to_int32_a_avx; + + if( out_fmt == AV_SAMPLE_FMT_FLTP && in_fmt == AV_SAMPLE_FMT_FLT || out_fmt == AV_SAMPLE_FMT_S32P && in_fmt == AV_SAMPLE_FMT_S32) + ac->simd_f = ff_unpack_6ch_float_to_float_a_avx; + if( out_fmt == AV_SAMPLE_FMT_FLTP && in_fmt == AV_SAMPLE_FMT_S32) + ac->simd_f = ff_unpack_6ch_int32_to_float_a_avx; + if( out_fmt == AV_SAMPLE_FMT_S32P && in_fmt == AV_SAMPLE_FMT_FLT) + ac->simd_f = ff_unpack_6ch_float_to_int32_a_avx; } if(HAVE_ALIGNED_STACK && channels == 8) { if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) |